Asymmetric Volatility Dynamics: Evidence From the Istanbul Stock Exchange

This paper considers estimating the conditional mean and variance from a single-equation dynamic model with the mean following an ARMA (1,7) process, and the conditional variance with time-dependent conditional heteroskedasticity as represented by ARCH models. The volatility is measured by a linear GARCH and an EGARCH process. Our results suggests that EGARCH provides better estimates than a linear standard GARCH model. The EGARCH also can capture most of the asymmetry, supporting the hypothesis that negative return shocks cause higher volatility than positive return shocks at the Istanbul Stock Exchange.